Munich Personal RePEc Archive

Quality of Internal Risk Rating

Frameworks at Commercial Banks in

Pakistan

Ali, Syed Babar

2012

Online at https://mpra.ub.uni-muenchen.de/55117/

MPRA Paper No. 55117, posted 09 Apr 2014 19:59 UTC

i

Quality of Internal Risk Rating Frawework at Commercial Banks in Pakistan

Syed Babar Ali

1

Abstract

The importance of internal risk rating system for an effective credit risk management system can

not be overemphasized. The system demands contunuous support and involement of top

management of the commercial banks, and the regulators. The attempt to develop robust internal

risk rating systems is ongoing among commercial banks. This study was made to measure the

quality of internal credit risk rating systems of commercial banks in Pakistan in terms of the

various aspects of an internal rating system.

To achieve this objective interviews of head of risk manageent of 10 commercial banks operating

in Paistan were conducted. The unstructured questions used in the interviews were transcribed

and the technique of content analysis was used. The findings revealed that internal risk rating

systems of commercial banks in Pakistan are generally strong but need improvement in a few

aspects. The following areas of internal risk rating systems of commercial banks in Pakistan

were found weak:

1. Environment specific to internal risk rating methodologies employed by the banks.

2. Environment specific to the documentation in the internal risk rating system.

Other areas such as credit grades, use of qualitative and quantitative factors, methodology,

internal risk rating policy, external credit rating, rating definitions, rating criteria, boards

involvement were found in lines with the internal risk rating guidelines issued by the SBP.

Key Words: Credit Risk, Credit Risk Management, Internal Risk Rating

2

Chapter 1 Introduction

1.1 Introduction to the Study

The title quality of internal risk rating framework at commercial banks in Pakistan is one having

high significance for the banking sector in Pakistan. This study is initiated at a time when

commercial banks are gearing strongly to face the challenges of this modern financial era. The

complexity, integration and the competitiveness of the financial markets have made it very

necessary for financial institutes and particularly commercial banks to manage their risks

effectively (Caouette et al., 2008, p.13). This study is about one of the most important area in

credit risk management which all of commercial banks are engaged into with firm commitment.

As the next chapter will explore more into it the internal rating system of a bank allow it to find

the credit risk inherent with each and every potential borrower. The system is very beneficial to

banks, needless to say; moreover the state bank of Pakistan has made it a requirement for every

commercial bank to maintain an internal rating system of a prescribed profile.

This study has been organized into six chapters such as: Introduction to the study, Introduction to

the Internal Risk Rating systems, literature review, research methodology, content analysis and

findings and conclusion.

The chapter introduction to credit risk deals with the environment and significance of internal

risk rating systems for commercial banks. The chapter is organized to first build the relationship

between credit risk, credit risk management and internal risk rating system. The remaining

3

section of the study discusses functions, expectations, developments and automation in the

subject matter of this study.

The third chapter Literature review deals with the discussion of the perspective of renowned

researchers in the area of internal risk rating framework practices. The chapter is divided into

areas such as benefits of Internal Rating System, Quantitative and Qualitative Factors, rating

models, Design of Internal Risk Rating , external and internal ratings, Basel Regulations and

Internal Risk Rating System, internal risk rating guidelines.

The description of internal risk rating best practices is based on the internal risk rating guidelines

issued by the state bank of Pakistan. The interpretation of the guidelines was very important, as it

allowed the researcher to determine the nature of the benchmark internal credit risk rating

framework and develop the interview questions accordingly.

The fourth chapter, mentioned as, research methodology, is about sampling plan and the research

techniques used in this study.

The fifth chapter, named as data collection and analysis, is separated into three parts. The first

part is a tabular presentation of data developed from the content analysis of the interviews made

with the heads of credit risk management. The second part is analysis of the data contained in the

tables. The third part, that is, findings, is composed of information derived from the analysis of

each individual table.

The final section of the study is the conclusion. The conclusion is derived from the output of the

content analysis combined with the insights developed from the review of the literature. The

main conclusion of the study was that internal risk rating framework of commercial banks in

4

Pakistan was generally good but needed improvement in certain areas such as credit culture,

credit monitoring, limit setting and credit risk modeling.

1.2 Problem Statement

The importance of internal risk rating system for an effective credit risk management system can

not be overemphasized. The system demands contunuous support and involement of top

management of the commercial banks, and the regulators. The attempt to develop robust internal

risk rating systems is ongoing among commercial banks. The commercial realize the coplexity in

the modern financial environment exposing them to even greater credit risk than before. Also,

establishing and maintaining an effective internal risk rating system is an uphill task in Pakistan

because of the lack of adequate default and and recovery data.

1.3 Objective of the Study

To measure the quality of internal credit risk rating systems of commercial banks in Pakistan in

terms of the following aspects of an internal rating system:

Credit grades, credit rating criteria, rating methodology, rating definitions, use of external

ratings, doocumentaion, reporting, use of quantitative and qualitative data and boards

involvement.

5

1.4 Type of study

This is a qualitative study made to evaluate the quality of internal credit rating systems of

conventional commercial banks. The quality of internal risk rating environment is measured in

terms of the specific set of guidelines issued by the state bank of Pakistan in 2008. The

instrument of in-depth interview was employed to gather data necessary to meet the aims of this

paper.

1.5 Scope and Limitation of the Study

The internal credit risk rating systems are established to grade a number of banking loans

including consumer, commercial and running finance loans. This study focuses on only the

commercial loans. Moreover, the research preview includes only conventional commercial banks

and ignores other financial institutes and Islamic banks. For the purpose of this research the set

of internal risk ratings guidelines issued by the state bank of Pakistan have been assumed as the

best practices in the area of internal risk rating.

6

Chapter 2 Introduction to the Topic

Most of the financial institutes today, and particularly commercial banks, maintain an internal

risk rating system to grade the credit quality of their customers. This is not an easy task rather

requires full commitment from the top management: commitment of physical and human

resources. The framework is an important input of credit risk management process as it allows

the bank to individually assess the credit risk of their customers. All financial intermediaries face

credit risk to varying degree of extent. Commercial banks, because of the nature of their

products, face credit risk at a higher level than is the case with other financial intermediaries

(Saunders & Cornett, p.173). Commercial banks, needless to say, are encouraged to sustain

highly effective systems to offset credit risk. This should be done by establishing and

maintaining an effective credit risk management system. Establishing an effective internal risk

rating system is a step in this regard; as, an internal rating system allows commercial banks to

measure and mitigate credit risks they are exposed to in their lending activities by aggregating

and managing the credit quality of the obligors (Bank of Japan 2005). Regulations governing

commercial banks have long appreciated the importance of credit risk management for

commercial banks which prompt them to monitor and regulate them very strictly.

2.1 Credit Risk

Credit risk is uncertainty associated with non-payment of a monetary obligation. There are three

types of credit risks such as default risk, down-grade risk and credit spread risk (Bessis, 2006,

p.13). Default risk is related to actual non-payment of obligation. Down-grade risk is the

7

probability that the credit rating will down-grade the issue or the firm. Credit spread risk is

associated with the probability that credit spread of the issue will decrease. To measure credit

risk effectively an active trading market should exists. As no active trading markets are available

for commercial loans generally and even so in Pakistan, the default risk becomes the most

relevant type of credit risk facing commercial banks in Pakistan.

2.2 Credit Risk Management

As mentioned above credit risk management is an essential component of the overall risk

management system of commercial banks. Credit risk management deals with identification,

measurement and mitigation of credit risk (Shimko and Went, 2010, p.40). It is at the second

step, measurement of credit risk, that internal risk rating system plays its part. It amounts to

having the ability and capacity to accurately risk rate each and every individual transaction and

customer. This was not practiced before as commercial banks were either relying on the data

received from the customer or using the risk rating of an external agency.

Basel II regulations deals extensively with credit risk and prescribe good practices for financial

institutes to measure credit risk. The methodologies prescribed in the Basel II regulations are

termed as standardized approach and Internal Rating Based approach (IRB) (Apostolik, et al,

2009, p.140-141). The regulations suggest the standardized approach for new banks and require

them to move forward towards the IRB approach. The essence of IRB approach is the existence

of a system allowing the banks to risk rate the borrowers by themselves. The internal rating

system then becomes the hallmark of the credit risk management system of the bank and allows

the bank to measure, price and manage credit risk of emanating from the individual customers

8

and the portfolio very effectively. Realizing the importance of internal risk rating the state bank

of Pakistan (SBP) issued a directive in 2008 in this respect. The directive asks the commercial

banks to establish and maintain an internal risk rating system to rate their borrowers. For

commercial banks to effectively manage it the state bank of Pakistan also released the internal

risk rating guidelines in 2008. The guidelines prescribe various aspects and procedures within the

internal risk rating systems of commercial banks. The guidelines focuses on the areas such as

credit grades, rating criteria, policy, architecture, rating methodologies and design, external

ratings, organizational structure and others. It is believed that the commercial banks need to

significantly adjust the level of computational and human resources to comply with the

expectations mentioned in the guidelines.

As mentioned above the implementation of internal risk rating in Pakistan is already underway.

The financial system, the regulator (SBP) and also the banks have realized the importance of

internal rating system. But the effectiveness of the system require full commitment from the top

management in terms of adhering to the essential principles of maintain the internal rating

system. In other countries the effectiveness of the system has, at times, been compromised

because of the issues such as lack of data (Servigny & Renault, 2004, p.48), inappropriate

modeling and lack of trained staff. The state bank of Pakistan has issued guidelines to support

the commercial banks to maintain an effective internal risk rating system. The guidelines are

comprehensive, detailed and at times flexible.

9

2.3 Internal Risk Rating

Credit risk is the most important exposure a financial risk faces in all of its business dealings.

Also, the style in which a bank measure and mitigates its credit exposures is extremely important

for its success. The erosion of equity through bad loans has seen many commercial banks

collapse over a period of time. Establishing and maintaining an effective internal risk rating

framework is the most important factor in managing the loan losses to the desired level.

Internal risk rating system is the framework which allows a commercial bank to rate their

customers in terms of their creditworthiness. The bank employs a rigorous process, starting with

the initiation of the loan application, to classify the transaction into a specific risk grade (Bessis,

2006, p.445). The grades are generally alphabetically named, such as (BBB, BB, B) or

numbered, such as (1-9) to quantitatively express credit risk.

In the words of Krahnen and Weber (2000) internal ratings is needed to support measurement of

credit risk, to effectively deal with the portfolio and to price the debts of the company. This

would only be meaningful if the internal ratings system employ strong controls and is

characterized by good practices. In a word with complex financial markets and transactions the

role of various external monitors is linked; auditors deal with the risk reporting systems of a

company, rating agencies analyze the risk measurement system of a prospective issuer. The

supervisory authorities, too, have started taking interest in the certification of various relevant

systems and models.

The system requires the bank to assess the credit risk of the transaction in terms of variety of

quantitative and qualitative factors. In the words of Sarac (2010) using specific criteria the

10

internal risk rating process produces credit score for each and every obligor. Then the scores are

aggregated in classes formed within a range of credit score interval. Each class is given a grade.

No single risk rating system is perfect in all conditions. Actually, the risk rating framework is a

function of the sophistication of the bank’s business activities. Large banks generally require

comprehensive rating systems with multiple grading to account for the complex level of credit

exposure arising from a wider product and customer base. The smaller banks need not be having

a very sophisticated rating system as the quality and depth of their interaction with their

customers allow them to incorporate more enhanced level of qualitative information in the credit

assessment process.

2.3.1 Functions of a Credit Risk Rating System

A well-management internal risk rating system promotes soundness and safety of the financial

institution. It produces categorization of individual loans into different risk categories (Sauders

& Allen, 2010, p.299). This requires a commercial bank to maintain a system to individually

measure the credit risk of each loan. The result is that commercial banks and the regulators can

be better aware of the changes in the credit risk and overall risk faced by the bank with the

facilitation of every new loan.

Internal credit risk ratings are very important for other critical functions as well. Some of these

functions are: credit approval, credit pricing, credit relationship, credit administration and

portfolio management information systems, and board reporting and portfolio management.

11

2.3.2 Expectations of Bank Credit Risk Rating Systems

One unique internal credit risk rating system can be considered as ideal for every situation. The

qualities narrated below should be the hallmark in every risk rating system, but banks can have

their own formula to merge those qualities to develop a process.

Following qualities are generally the part and parcel of risk rating systems:

The rating system should work in the integrated environment, promote measuring, monitoring,

and communicating of credit risk, and provide important inputs to the strategic decision making

framework of the bank. Duly approved by the board the risk rating system should work in an

integrated environment. That way it will promote credit monitoring, measurement and reporting.

The best of all it will support the board decision making. The board should also set responsibility

and accountability for the framework. Sufficient information should be provided to the

management for effective implementation. The risk rating process should be applied to all of the

exposures. The number of ratings should be adequate. Risk ratings should be precise and be

contemporary in nature. The criteria for risk rating should have clarity and precision and defined

based on both quantitative and qualitative factors. In terms of risk representation the ratings

should account for both the borrower’s anticipated behavior and the nature of the transaction. It

should not be static, and be independently validated. The rating assigned to an exposure should

be adequately documented. .

12

2.3.3 Developments in Bank Risk Rating Systems

Numerous banks outside Pakistan are creating strong internal risk rating processes so as to

improve the precision and robustness of credit risk management system. This trend is likely to

continue as banks employ sophisticated diversification approaches and strengthen the systems to

measure economic and regulatory capital to credit risk. Moreover, more compliance is in order

for commercial banks that follow the Basel Committee recommended internal-ratings-based

approach ( Shimko and Went, 2010, p.137-138). Increasingly banks are:

Increasing the number of credit grades they employ; using dual rating systems; using risk rating

systems to measure default and loss probabilities; and using models in the rating system.

2.3.4 Automated Scoring Systems

Previously automated systems were only used for consumer loans and corporate bonds. But with

the availability of data and the advancement in information technology has come the possibility

of using an automated system in the internal rating system. The commercial banks employ these

systems to a range of activities in the system. This depends on the type of the automated system

and the architecture of the rating system. Whatever is the level of involvement of the automated

system the bank has to fulfill the requirements of the guideline and not refer to the vendor’s

claims as a way of meeting the requirements of the guidelines.

The models employed are characterized as either quantitative or expert systems. A statistical

system measures credit risk based on the quantification of factors identified by the vendor as

representation of the credit exposure. An expert system, on the other hand, mimics an analyst’s

style in the credit decision.

13

Chapter 3 Literature Review

3.2 Benefits of Internal Rating System

Risk rating is beneficial because it allows banks to improve in many dimensions. It requires

credit officer to make a credit assessment of an application before it is approved. It allows the

department to measure the overall credit risk of the portfolio helping them to know whether or

not the bank can tolerate further risk. Then it allows the bank to quantitatively measure the

impact of the credit decisions on the overall risk of the bank. Taken together these advantages

allow a bank to price its loan more accurately. According to English and Nelson (1998) all banks

strengthen their credit risk management frameworks by the employment of internal ratings

system.

English and Nelson (1998) Internal risk rating is very common among all banks with larger

banks having more complicated systems than the smaller ones. Also, larger banks maintain more

risk categories than the smaller banks.

One more benefit is that monitoring activities could be allocated more judiciously. The loans

having lower ratings can be monitored more stringently. The widespread application of internal

risk rating would provide liquidity to the business loans, ability to the bank to measure loan loss

more accurately thus resulting in the overall boost to the economy.

14

3.3 Quantitative and Qualitative Factors

For the borrowers internal risk ratings, credit grades are generally determined in terms of the

assessment employing both quantitative and qualitative factors. Bina Lehmann (2003) is not

fully sure whether the additional costs incurred to obtain qualitative information is justified but

opined that judgments are very important and add value to the internal rating frameworks.

Comptroller’s handbook (2001) state that the internal rating process begins with a

comprehensive assessment of the obligor’s based among other factors on judgmental factors.

Substantiating the usefulness of qualitative factors, Bank of Japan (2005) states that in case

quantitative factors cannot precisely measure the credit risk, the impact of qualitative factors

should be incorporated. Specifically, such an impact should be used to either adjust the credit

score or the credit grade. The employment of qualitative factors can also make the whole process

very subjective and fraught with inconsistencies. That is why Bank of Japan (2005) suggests that

detailed evaluation criteria are necessary for qualitative assessment, which should be as specific

as possible and well documented.

Similarly Treacy and Carey (2000) state that the human judgment of experienced staff is very

important in the assignment of ratings, so much so that banks maintain the operating design

which promote accuracy and consistency of ratings but also do not restrain the judgment.

3.4 Rating Models

Bina Lehmann (2003) is not fully sure whether the additional costs incurred to obtain qualitative

information is justified but opined that judgments are very important and add value to the

15

internal rating frameworks. Comptroller’s handbook (2001) state that the internal rating process

begins with a comprehensive assessment of the obligor’s based among other factors on

judgmental factors.

3.5 Design of Internal Risk Rating

According to Monica and Monica (2009) the important components of the internal rating system

are: a procedure of creating the parameters, well-defined categories, a responsibility structure of

the process, physical resources, quantitative models incorporating the qualitative factors, and

validation.

Treacy and Carey (2000) state that the important considerations in the design of the system

include the responsibility for grading, the reviews of ratings, authority structure, the agency

ratings, quantitative models, the formality of the process and rating definitions.

Frerichs and Wahrenburg (2003) state that financial institutes cannot improve system quality if

they do not develop rating classes based on the aggregation of credit scores. He further stated

that this also results in increase in capital because of the inherent non-linear nature of the capital

function.

Pascal Damel (2006) Linked the internal rating framework, development in credit derivatives

models, and the development of sophisticated credit risk measurement models.

16

Kauko stated that the through-the cycle ratings are believed to reduce the procyclicality of the

credit risk. But added that their result based on Moody’s KMV data on Finnish firms shows

negative result as companies did not seem to follow credit risk cycles consistently.

Jacobson, Jesper, Roszbach find that design, other parameters and the implementation are a few

controls which effectively allow a bank to account for its credit risk and therefore allow it to

operate without additional layer of capital. Their research, though, showed contrasting result.

According to Treacy and Carey (2000), design of the rating system are based on the nature of

bank loans, relevant cost, staff which uses the ratings and the importance of ratings in the

development of desirable type of credit culture.

3.6 External and Internal Rating

Nakamura, Roszbach and Riksbank (2010) report that while assigning ratings information is lost,

which implies that it is optimal to combine important information derived from external ratings

with the internal credit ratings framework to measure credit risk.

Yueh and Webber (2003) report that as compared to external ratings the internal rating system is

more beneficial to a bank as it allows bank to efficiently respond to changes in credit qualities.

This is possible because it incorporate private information as well as the judgment of the bank.

Comptroller’s handbook (2001) states that external ratings provide one perspective of a

customer’s credit risk; therefore, the assessor risk rating must be based on his own assessment of

17

the environment facing the customer. The report also adds that if the banks have input from

external ratings into their risk rating framework then they must adjust their internal ratings when

credit risk changes, whether or not the external ratings have remained unchanged.

3.7 Basel Regulations and Internal Risk Rating System

Mendoza (2005) The IRB approach allows banks’ to use internal ratings of their loans to

measure credit risk. All exposures are separated into categories using prescribed criteria and

descriptions. While effectiveness of IRB approach has been cited by many researchers still there

are points of criticism from a number of researchers. Varotto (2008), for example, states that the

internal rating based approach (IRB) to measure credit risk under Basel II regulations considers that the

portfolio is not concentrated and a single factor describes risk. Their test with a general credit risk model

showed that differences in the IRB and the general model were significant.

3.8 Internal Risk Rating Guideliness

Comptroller’s handbook (2001) provide the following guidelines to maintain an effective

internal risk rating system:

The risk rating system need to gel with the other facets of the credit environment so as to

enhance the management decision making capabilities. The role of board is also prescribed; it

should not only approve the policy but also set responsibility structure dealing with the internal

risk rating framework. Sufficient information should be provided to the management for

effective implementation. The risk rating process should be applied to all of the exposures. The

18

number of ratings should be adequate. Risk ratings must be accurate and timely. The criteria for

risk rating should have clarity and precision and defined based on both quantitative and

qualitative factors. The ratings should be a representation of the risks faced because of the

obligor’s anticipated performance and the nature of the loan. The system need to be dynamic and

independently validated. The rating assigned to an exposure should be adequately documented. .

3.8.1 Scope of Ratings:

The commercial banks in Pakistan are bound by the regulations to credit rate all of type of their

banking loans. The ratings for commercial loans, which are the probability of default, should be

based on the risk of borrower default.

3.8.2 Rating grades/structure:

The internal risk rating systems of the banks should have an adequate number of risk grades so

that concentration of risk grades is not excessive.

The commercial banks should maintain adequate no of risk grades to have a meaningful

distribution of risk. The no of credit grades should not be less than nine for loans without default

and three for loans which have already defaulted. Rating definitions, procedures and criteria for

attributing exposures to the grades must be precisely described. Written rating definitions must

be specified clearly to let third parties to understand and mimic rating assignments and perform

evaluation of the grades.

19

All of the banks must establish a methodology that distinguishes meaningfully between the credit

worthiness of sequential credit grades. The specifications must be in such details as to ensure

consistency in the assignments. This consistency should occur across every relevant dimension

of the business. A bank must state in its credit policy the level of risk each risk grade implies.

Risk perception must increase with the decline from one grade to the next. The risk of each grade

should be described in terms of the probability of default risk and the criteria used to characterize

that level of risk

3.8.3 Rating criteria:

5.1 In terms of rating criteria the guidelines prescribe that to assign ratings effectively the banks

need to incorporate the impact of all related and significant information. All of the important

factors related to the borrower should be addressed. Any factor specific to the financial

conditions and managerial domain of this business cannot be ignored. Similarly, all of the

important quantitative and qualitative factors should be considered.

The financial condition should be assessed with regard to the level of debts and financial

performance such as earnings and levels of cash flows. The management structure should be

evaluated based on ownership structure, management quality, willingness and strength of

financial position. The qualitative factors should be evaluated based on the nature of CIB report,

business sector and industry characteristics.

20

3.8.4 Rating assignment horizon:

In terms of the rating horizon, the guidelines prescribe one year but recommend a higher period

than one year. This is required because the borrower rating should represent the financial

conditions that the borrower will face and its willingness to pay over a long-term period. The

long-term period would see varying degrees of financial conditions that a borrower is likely to

face. The assignments could be subjected to stress testing. The other option could be to

incorporate the impact of adverse financial conditions without involving stress testing.

3.8.5 Use of External Ratings:

The guidelines prescribe that with the availability of external credit risk rating the banks need to

make sure that all relevant risks have been incorporated in the internal risk rating. In case the

external rating is very different from the existing internal rating, an explanation is in order. The

guidelines recommend negative adjustment but disallows positive adjustment.

3.8.6 Rating migrations/back testing:

For an effective credit risk management system, it is inevitable to continually monitor the credit

worthiness of the obligor and the worth of collateral to determine the actual level of risk.

Therefore it is important that borrowers’ ratings are reviewed at least annually. Higher risk

obligor, needless to say, must have their ratings reviewed more frequently. The bank should have

an accurate and quick process to gather and incorporate significant information related to the

21

borrower. Once new information is received, the systems available at the banks must be able to

incorporate it with the existing knowledge to arrive at the new rating without much delay.

The guidelines recommend that the risk rating system should be concurrent with a powerful

information system so as to effectively facilitate the bank in monitoring actual default rates. It is

also prescribed that the banks maintain the relevant credit data to perform validation.

Furthermore, it is suggested to banks to maintain a procedure to back-test and recalibrate the

assigned ratings.

The banks should maintain a historical data specific to the changes in the assigned ratings and

process it into objective information. The banks need to also determine the name of the factors

influencing the changes in credit ratings, and use these factors to continue to strengthen the

systems. The banks should have historical data related to the ratings. The information such as the

date of the original assignment, date of the revision, the staff who were involved in the rating, the

methodology employed and the nature of the data used should be part of the records. The

commercial banks should also make sure that data related to the defaults is recorded

appropriately.

3.8.7 Documentation of rating system design:

Consistency of application is a very important aspect of a rating system. This is attained with

adequate level of documentation and the training of the staff involved in rating.

The design and details of the rating system must be documented by the banks. It must also

account for the rating systems areas such as authorities of staff, oversight by the management,

22

criteria, diversification and periodicity of reviews. The reasons for the adoption of a particular

criteria and analyses practically explaining the criteria should also be documented.

It is also very important that the bank record the historical data related all significant changes

made in the rating system. In case the bank uses a quantitative model the documentation should

address the functional specifications of the model as well.

3.8.8 Corporate Governance and oversight:

It is the board obligation to approve an internal risk rating policy. The policy could be separate

from the credit risk policy, if that is the case, then It should be approved as such. Otherwise the

approval of the credit policy would be deemed as the approval of the internal risk rating policy as

well. The board should maintain an adequate and consistent level of vigil over the rating system.

The policy should be reviewed at regular intervals to determine its relevance with the portfolio

and economic conditions. The guidelines prescribe that the regular reporting of bank’s portfolio

quality should include information about risk ratings. It further suggests the inclusion of analysis

of credit grades in terms of segments and portfolios. Further information would be the analysis of

historical default rates and migration analysis.

Banks are required to operate credit risk control activities which are accountable for the

effectiveness of the internal rating systems. The department must operate independently from

other functions such as origination. Furthermore it should test and monitor grades, implement

processes to vouch for the consistency of definitions, perform reviewing and record variations. It

23

must also be responsible to supervise and overview the usage of any models employed. Rating

assignments and reviews must be done or sanctioned by a party which would not create a conflict

of interest. This independence can be accomplished by a number of approaches. The approach

adopted by the bank should be documented and made part of the credit policy of the bank.

3.8.9 Reporting Requirements:

The banks have a requirement to possess an internal risk policy fully endorsed by the board and

establish fully functional internal risk ratings systems. The criteria to form ratings should be

transparent and without ambiguities. The banks are required to report nine grades for non-

faulting loans and three grades for default categories to the state bank of Pakistan. The reporting

to the directors should include credit grade in terms of portfolios, financial products,

geographical locations; the reporting should also include credit migration and actual loss rates.

24

Chapter 4 Research Methodology

4.2 Sampling Method

There are about 29 commercial banks operating in Pakistan including 6 Islamic banks. Excluding

the Islamic banks ten commercial banks were selected based on the convenience of the

researcher.

4.3 Research Technique

Interviews of the head of credit risk management were conducted. The unstructured form of

interview was used to capitalize on the expert opinions of the head of risk management or head

of credit risk management. The questions employed in the interviews were framed based on the

contents of the internal risk rating guidelines issued by the state bank of Pakistan in 2008.

The transcription of the interviews was performed. The transcriptions were subjected to content

analysis. The information in the transcription was sorted in terms of various aspects of the

internal risk ratings such credit grades, rating criteria, rating definition, documentations, internal

risk rating policy and relationship with external credit ratings. Then categories were formed for

each of the aspects where each category represented a unique piece of information on the

individual aspect.

25

Chapter 5 Data Collection and Analysis

5.1 Introduction

We asked a total number of 20 questions to the head of risk management of 10 commercial

banks operating in Pakistan. These open ended questions resulted in diverse responses from the

interviewee. We classified these responses into 3 to 5 categories depending on the number of

different responses obtained on the question asked. The categories were formed based on the

similarity of the responses from the interviewees. As the questions were open-ended each

interviewee could have a response featuring in more than one category. For each of the questions

a tabular presentation was made that exhibited the response of each of the commercial bank in

terms of the categories formed.

An overall analysis of the responses from all of the commercial bank was given for the tabular

presentation of each of the question asked. This analysis represented the raw data for the credit

risk management practices by commercial banks in Pakistan.

26

5.2 Content Analysis

5.2.1 Credit culture of the bank.

Category 1 Consistency of practices with risk appetite & credit policy

Category 2 Strong Management at the Top of Credit Function

Category 3 Clear Accountability of Every Personnel Involved Category 4 Regular training on the credit policy and credit analysis

Category 5 Reward for Vigilance and Penalty for Negligence

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √

2 √

3 √ √ √ √ √

4 √

5 √ √ √ √ √

6

7l √

8 √ √

9 √ √

1 √ √

Based on the content analysis of the elite interviews conducted with the head of credit risk/head

of risk management it is found that credit culture among the commercial banks is generally

weak. While 70% of the respondents say that credit practices are in line with the credit poly and

risk appetite only 50% considered the top management of credit function as being strong.

Another alarming aspect of the environment specific to the credit function was poor quality of

training provided to the management and staff. When the issue of incentive was discussed only

30% sated that adequate system of reward and penalty was in place to motivate vigilant

employees. Similarly only 40% stated that clear accountability framework was in place to hold

concerned employees accountable.

27

5.2.2 Credit Grades

Category 1 The no of grades reflect a meaningful distribution of exposures

Category 2 The no of grades do not cause excessive concentrations

Category 3 The bank has more than 9 grades for non-defaulted borrowers Category 4 The bank has three grades for defaulted borrowers

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √

2 √ √ √

3 √ √ √ √

4 √ √ √

5 √ √ √ √

6 √ √ √

7l √ √ √

8 √ √ √

9 √ √ √

10 √ √ √

All of the 10 commercial banks interviewed had credit grades available which did not create

excessive concentration of credit risk in the loan portfolio. Also the structure of the grades was

so framed in terms of the number and range of grades that created a smooth distribution of risk

definitions.

Most of the banks, that is 8 of them, had 9 grades for non-defaulting loans; two had more than 9.

For defaulting loans all of the 10 banks had 3 grades.

28

5.2.3 Rating criteria:

In assignments of ratings the banks take into account the following factors related to the

borrowers:

Category 1 Financial condition in terms of debt burden, earnings and cash flows.

Category 2 Management analysis, systems of internal controls, payment performance,

and financial condition of the sponsors.

Category 3 CIB report, sector of business and industry analysis

Respondent Category 1 Category 2 Category 3

1 √ √ √

2 √ √ √

3 √ √ √

4 √ √ √

5 √ √ √

6 √ √ √

7l √ √ √

8 √ √ √

9 √ √ √

10 √ √ √

In terms of the consideration of relevant borrowers’ factors the internal risk rating framework

maintained by the commercial banks in Pakistan was found effective. All of the ten banks

interviewed. The ten banks interviewed had mixed views about the attributes of the borrowers

that they assess to assign a rating. All of the banks expressed their complete confidence over the

assessment of financial conditions such as the debt burden, earnings potential and the expected

levels of cash flows of the business. About the various factors associated with the analysis of the

management of the business the bankers had different views. No all of them thought that analysis

of management, controls in the business, payment performance and sponsors financial strength

were the factors that they always considered. In fact only 50% of the banks were in favor of

these factors. About the qualitative factors such as information in the CIB report and industry

analysis all of the banks stated that these were important.

29

5.2.3 Documentation of Internal Risk Rating Framework

Category 1 Portfolio differentiation

Category 2 Rating criteria

Category 3 Responsibilities of parties that rate borrowers

Category 4 Frequency of rating review

Category 5 Management oversight of the process

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √

2 √ √ √ √

3 √ √ √ √

4 √ √ √

5 √ √ √ √

6 √ √ √

7 √ √ √

8 √ √ √

9 √ √ √

10 √ √ √ √ √

The interviewees revealed very important information about the documentation of the internal

risk rating process. The discussion revealed issues specific to documentation of portfolio

differentiation, rating criteria, responsibilities of the staff, frequency of ratings and management

oversight of the process.

Majority of them, that is, 9 out of 10, stated that documentation was adequate in terms of

portfolio differentiation. In terms of rating criteria, the interviews revealed that, 8 out of 10 had

effectively documented the rating criteria. The documentation for the responsibilities of the staff

involved in rating was not so impressive for most of the banks as only 4 out of 10 were found

effective in this regard. The frequency of the rating reviews was well recorded by the banks but

documentation of the management oversight process had problems. 6 out of 10 banks seemed to

have effective documentation the other 4 did not have effective documentation in this respect.

30

5.2.4 Development of the internal risk rating system? If so, please describe it.

Category 1 Yes, and using to measure credit risk

Category 2 Has recently done so

Category 3 The bank has acquired a vendorised analytic system

Category 4 Finding It Difficult to Implement the Analytic

Respondents Category 1 Category 2 Category 3 Category 4

1 √ √

2 √ √

3 √ √

4 √ √

5 √ √

6 √ √

7 √ √

8 √ √

9 √ √

10 √ √

Existence of Internal risk rating framework is necessary if the banks desire to implement the

advanced approach to measuring credit risk. All of the banks have acquired a vendorised system

to determine the inputs to measure credit risk. Overall 70% of the banks stated that they had the

internal credit rating system to measure all the components of credit risk. 20% stated that they

had just been able to establish the internal credit rating system. Only one bank stated that they

had recently acquired the vendorised analytic system so they were having some difficulties

establishing the internal credit rating system

31

5.2.5 External credit ratings

Category 1 The available external credit ratings are considered while finalizing the

internal ratings.

Category 2 The internal risk rating is not completely aligned with the external rating

Category 3 An explanation is produced when an external rating is quite different from

the internal rating

Category 4 In accordance with the movement in external credit rating the internal risk

rating is only adjusted downward and not upward

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √ √

2 √ √ √ √ √

3 √ √ √ √ √

4 √ √ √ √ √

5 √ √ √ √ √

6 √ √ √ √ √

7 √ √ √ √ √

8 √ √ √ √ √

9 √ √ √ √ √

10 √ √ √ √ √

The interviews revealed that all of the 10 banks considered external credit ratings while

finalizing the internal ratings. The external considered were considered but not totally aligned

with the final outcome of the assignments. The internal rating systems of all of the banks

required an explanation if the internal ratings diverged significantly from the external ratings.

Also, all of the banks the consideration of external ratings was limited to only revising the

internal ratings downward.

32

5.2.6 Rating Definitions

Category 1 Rating definitions and criteria for assignment of exposures are precisely

described

Category 2 Rating definitions allow internal auditors to understand assignment

Category 3 Rating definitions allow third parties to replicate assignments

Category 4 Ratings result in meaningful differentiation of risk

Respondents Category 2 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √ √

2 √ √ √ √

3 √ √ √ √

4 √ √ √ √

5 √ √ √ √

6 √ √ √ √

7 √ √ √ √ √

8 √ √ √ √

9 √ √ √

10 √ √ √ √

The interviews with the head of risk management revealed that all of the banks clearly had

clarity in the definitions of rating grades and the criteria for assigning these was also clearly

defined. In terms of the replication of the ratings by internal auditors only 6 banks out of 10 were

found to have this attribute. The replication of the ratings by third parties had the similar

response from the banks, only 5 out 10 banks were found to maintain clarity to that level. Lastly,

all of the banks were found to have meaningful definitions of rating to achieve differentiation of

the credit risk.

33

5.2.7 Internal Risk Rating Reporting

Category 1 The reporting to the board include rating information

Category 2 Reporting of credit rating information is broken down by grades.

Category 3 Reporting includes analysis of realized default rates

Category 4 Reporting includes trend and migration analysis

Category 5 Reporting include information on segment breakdown by credit grade

Respondents Category 1 Category 2 Category 3 Category 4 Category 5 Category 6

1 √ √ √ √

2 √ √ √ √

3 √ √ √ √ √

4 √ √ √ √

5 √ √ √ √ √

6 √ √ √ √ √

7 √ √ √ √

8 √ √ √

9 √ √ √ √ √

10 √ √ √ √ √

The internal risk rating reporting practices among commercial banks were found to be varying to

a certain extent.

All of the banks interviewed stated that the periodical reporting to the boards included rating

information. Also, all of the banks stated that the information was broken down in terms of credit

grades. This was contrasted by the fact that only 40% of the banks sated that the information

included realized default rates. In terms of the reporting of trend and migration analysis only

50% of the banks sated that they did so.

The frequency of the reporting was also another area where divergence of practice was found.

50% of the bank stated they were reporting credit information quarterly, that is, for each meeting

quarterly meeting of the directors. The remaining banks sated that they were reporting the

information annually.

34

5.2.8 Board Involvement in the System

Category 1 Board of directors approves the internal risk rating policy.

Category 2 Top management approves the internal risk rating policy

Category 3 Internal risk rating policy is approved as part of credit policy

Category 4 Internal risk rating policy is approved separately

Category 5 Board set the responsibility structure

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √

2 √ √ √

3 √ √ √

4 √ √ √

5 √ √ √

6 √ √ √

7 √ √ √

8 √ √ √

9 √ √ √

10 √ √ √

All of the banks interviewed stated that they had internal risk rating policy sanctioned through

the board of directors. Out of the 10 banks interviewed 8 said that they had the policy approved

as part of the overall credit policy. The remaining 2 banks stated that the internal credit risk

rating policy was separately sanctioned by the directors.

In terms of the board’s involvement in the setting of the responsibility structure all ten banks

were found to experience the requisite role of the board.

35

5.2.9 Internal risk rating framework linkages to the systems of the Banks.

Category 1 Portfolio monitoring,

Category 2 Analysis of loan loss reserves

Category 3 Loan pricing

Category 4 Capital allocation and analysis of return

Category 5 A single rating system for lending, risk measurement and allocation

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √ √

2 √ √ √ √ √

3 √ √ √ √ √

4 √ √ √ √ √

5 √ √ √ √ √

6 √ √ √ √ √

7 √ √ √ √ √

8 √ √ √ √ √

9 √ √ √ √ √

10 √ √ √ √ √

All of the banks interviewed disclosed that they were using internal risk rating system for

numerous purposes. They were found to use it for portfolio monitoring, loan loss provisioning,

loans pricing and capital planning.

This implied strong that the banks did not have separate systems of credit ratings for various

credit risk management activities.

36

5.2.10 Internal Risk Rating Methodology

Category 1 Methodology is based on size, nature of operations and clientele base.

Category 2 Methodology is flexible and accommodates current and prospective risks

Category 3 Methodology accommodates estimated degree of diversification

Category 4 Methodology accommodates complexity of lending transactions

Category 5 Methodology provides effective information for credit risk management

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √

2 √ √ √ √

3 √ √ √ √ √

4 √ √ √ √ √

5 √ √ √ √ √

6 √ √ √ √ √

7 √ √ √

8 √ √ √ √ √

9 √ √ √

10 √ √ √ √ √

All of the banks interviewed have been found to maintain a structure methodology to assign

ratings to obligors. All of the banks were found to have methodologies which were based on the

size and nature of operations as well as the clientele base. In terms of risks, 60% of the banks

were perceived to have methodologies that were found to be flexible enough to accommodate

both current and prospective risks. 40% of the banks were found to be lacking a methodology

that could accommodate both current and prospective risks. The aspect of anticipated

diversification was also accommodated by the banks, as all of the banks were perceived to do

this. It was also found that the methodology were in line with the sophistication of the

commercial lending activities.

The banks were believed to maintain methodologies which were also producing information that

was adequate and sufficient for the maintenance of an effective system of credit risk

management.

37

5.2.11 Use of quantitative and qualitative factors

Category 1 Quantitative factors

Category 2 Qualitative factors

Category 3 Both Quantitative and qualitative factors

Category 4 Relies heavily on quantitative factors

Category 5 Relies heavily on qualitative factors

Respondent Category 1 Category 2 Category 3 Category 4 Category 5

1 √ √ √ √

2 √ √ √

3 √ √ √

4 √ √ √ √

5 √ √ √

6 √ √ √ √

7 √ √ √

8 √ √ √ √

9 √ √ √

10 √ √ √ √

All of the banks interviewed stated that they used both quantitative and qualitative factors to

assign credit ratings.

Two of the banks were found to rely heavily on quantitative factors whereas three were found

that rely heavily on qualitative factors. According to those relying heavily on qualitative factors

judgment of the underlying situation facing the employer was more important than the

quantitative measure of the credit score.

38

Chapter 6 Findings

6.1 Introduction

We interviewed head of risk management of 10 commercial banks in Pakistan and used the

technique of content analysis to obtain desired information. The information we obtained is

presented below under the headings of credit culture, credit review, limit setting, problem

credits, credit reporting, credit approval, stress testing, monitoring, pricing credits, collateral

policy and credit risk modeling.

6.2 Credit Grades

The internal risk rating framework maintained by the commercial banks in Pakistan is effective

in terms of the nature of credit grades maintained within the framework. The banks have right

number of grades which does not create credit risk concentration and provide a functional

distribution of risks. No of credit grades required by the SBP guidelines is 9 for non-defaulting

and 3 for defaulting loans, all of the banks in Pakistan has at least this no of grades.

6.3 Use of quantitative and qualitative factors

Another important aspect of internal risk rating framework is the use of quantitative and

qualitative factors in assessing the credit risk of the borrower. The SBP guidelines suggest a

mixture of the factors which would appropriate given the financial economic condition faced by

the borrower. The banks in Pakistan are found to have a mix of qualitative and quantitative

factors to suggest a full compliance with the internal risk rating guidelines.

39

6.4 Internal Risk Rating Methodology

The commercial banks in Pakistan were found to have not so robust methodologies employed in

their risk rating framework. The methodologies were found to be adequate in terms of the size

and nature of lending activities. But what they lacked in was flexibility to accommodate future

risks profile and anticipated diversification. The methodologies, however, were found to be

sophisticated enough to perform under modern lending environment.

6.5 Documentation

The commercial banks in Pakistan were found to have weaker documentation environment to

cater to different activities in the internal risk rating system. Although the documentation was

adequate in terms of portfolio differentiation and not so bad for rating criteria, it did not do well

in terms of other aspects of internal risk rating framework. The documentation for the

responsibilities of the staff and the management oversight process were found weak.

6.6 Internal Risk Rating Policy

All of the banks interviewed were found to have good internal risk rating policy practices in

terms of its approval and revisions. They banks were found to have the policy approved from the

board of directors - a good practice in terms of the SBP guidelines. The revisions were made by

the bank annually or whenever the economic or financial circumstances demanded so; again a

good practice in terms of the guidelines.

6.7 External Credit Ratings

The interviews revealed that all of the commercials considered external credit ratings before

finalizing the internal ratings. Although this was done but total alignments between the two types

40

of the credit ratings was not sought by all of the commercial banks. The internal risk rating

framework of all of the banks required an explanation if the internal ratings diverged

significantly from the external credit ratings. Also, for all of the banks the consideration of

external ratings was limited to only revising the internal ratings downward.

6.8 Rating Definitions

The interviews with the head of risk management revealed that all of the banks had clarity in the

definitions of rating grades and in the criteria for assigning the internal risk ratings. In terms of

the replication of the ratings by internal auditors only 6 banks out of 10 were found to have this

attribute. The replication of the ratings by third parties had the similar response from the banks,

only 5 out 10 banks were found to maintain clarity to that level. Lastly, all of the banks were

found to have meaningful definitions of rating to achieve differentiation of the credit risk.

6.9 Rating Criteria

The ten banks interviewed had mixed views about the attributes of the borrowers that they assess

to assign a rating. All of the banks expressed their complete confidence over the assessment of

financial conditions such as the debt burden, earnings potential and the expected levels of cash

flows of the business. About the various factors associated with the analysis of the management

of the business the bankers had different views. No all of them thought that analysis of

management, controls in the business, payment performance and sponsors financial strength

were the factors that they always considered. In fact only 50% of the banks were in favor of

these factors. About the qualitative factors such as information in the CIB report and industry

analysis all of the banks stated that these were important.

41

Chapter 7 Conclusion

Credit risk comes from the expectation about borrower’s perceived inability and unwillingness to

pay. So a financial institute is a unique seller as it has to select and monitor its customers unlike

other types of businesses. The deficiency in the selection process or the monitoring process can

be disastrous for the financial institute even when the financial product is in great demand.

Commercial banks are inherently exposed to this risk, and only an effective credit risk

management system can protect them from undesirable losses. This is particularly important for

a country like Pakistan where commercial banks are limited by the lack of good governance,

constraint of physical and human resources.

The literature on internal risk rating system practices employed by commercial banks is not very

comprehensive and rich. The most popular aspect of internal risk rating systems among the

researchers has been quantitative and qualitative factors, rating models, design of the internal

rating systems and external and internal ratings. According to the researchers to operate

effectively the internal risk rating systems need to attend to to these aspects.

The internal risk ratings guideliness were issued by SBP in 2008. The contents of the

guideliness, that is, risk rating principles and the requirement imposed by the SBP were used as

the basis for evaluating the quality of risk rating practices employed by the cmmercial banks in

pakistan.

42

In this study we attempted to find out the quality of internal risk rating systems prevalent with

the commercial banks in Pakistan. To achieve this objective interviews of head of risk

manageent of 10 commercial banks operating in Paistan were conducted. The unstructured

questions used in the interviews were developed in accordance with the key attributes of the

internal risk rating guideliness issued by the state bank of Pakistan. The interviewd were

transcribed and the technique of content nalysis was used to obatin common patterns in the

responses. The findings revealed that internal risk rating systems of commercial banks in

Pakistan are generally strong but need improvement in a few aspects. The following areas of

internal risk rating systems of commercial banks in Pakistan were found weak:

Environment specific to internal risk rating methodologies employed by the banks.

Environment specific to the documentation in the internal risk rating system.

Other areas such as credit grades, use of qualitative and quantitative factors, methodology,

internal risk rating policy, external credit rating, rating definitions, rating criteria, boards

involvement were found in lines with the internal risk rating guidelines issued by the SBP.

Therefore these aspects of internal risk rating systems at commercial banks in Pakistan could be

considered strong.

43

References

Apostolik et al., 2009. Foundations of banking Risk. Hoboken: Global Associations of Risk Professionals in association with John Wiley & Sons, Inc.

Bank of Japan, 2005. Advancing Credit Risk Management through Internal Rating Systems. Tokyo:Risk assessment

Section

Basurto, M. S. and Lowe, P.W., 2002. Internal Ratings, the Business Cycle and Capital Requirements: Some

evidence from an Emerging Market Economy. BIS Working Paper No. 117.

Bessis, J., 2006. Risk Management in Banking. 2nd ed. West Sussex: John Wiley & Sons Ltd.

Caouette et al., 2008. Managing Credit Risk: The Great Challenges for Global Financial Markets. 2nd ed. Hoboken:

John Wiley and Sons, Inc.

Damel, P., 2006. How to calculate an internal rating? Synthesis and Proposition for a Contingent Approach using a

Monte Carlo simulation and a rate stochastic model following poisson’s law. International journal of business, vol.

11,no.2.

English, W.B. and Nelson, W.R. 1998. Bank Risk Rating of Business Loans, working paper , Federal Reserve

Board.

Frerichs, H. and Wahrenburg, M., 2003. Evaluating internal credit rating systems depending on bank size. Paper

presented at the 6th conference of the Swiss society for financial markets research, EFMA, Frankfurt.

Hertig, G., 2005. Using Basel II to Facilitate Access to Finance: The Disclosure of Internal Credit Ratings. ECGI -

Law Working Paper No. 31/2005.

Jacobson, T., Linde, J. and Roszback, K. F., 2006. Internal ratings, implied credit risk and the consistency of banks’ risk classification policies. Journal of Banking & Finance, vol. 30(7), pp.1899-1926.

Kauko, K., 2012. Can credit risk be rated through-the-cycle? Frontier in Finance and Economics, vol. 9, p.1-32.

Kirstein, R., 2001. The New Basle Accord, Internal Ratings, and the Incentives of Banks. International Review of

Law and Economics, Vol. 21, No.4.

Krahnen, J.P. and Weber, M., 2000. Generally Accepted Rating Principles: A Primer. Journal of Banking and

Finance

Lehmann, B., 2003. Is it worth the while? The relevance of qualitative information in credit rating. EFMA 2003.

Helinski meetings.

Mingo, J. and Falkenstein, E., 2000. Internal Credit Risk Rating Systems Must Evolve to Stay Relevant. Journal of

lending and Credit Risk Management.

Monica, D. and Monica, B.M., INTERNAL RATINGS SYSTEMS: AN EMPIRICAL APPROACH

Sarac, M., 2010. Measuring the effectiveness of internal credit rating: an application on a participation bank. International Research Journal of Finance and Economics. EuroJournal Publishing, inc.

44

Saunders, A. and Cornett, M. M., 2008. Financial Institutions Management: A Risk Management Approach. 6th ed.

USA: The McGraw-Hill/Irvin.

SBP., 2010. Guidelines on Internal Credit Risk Rating Systems. Karachi:

Saunders, A. Allen, L., 2010. Credit Risk: Measurement In and Out of the Financial Crisis. 3rd ed. Hoboken: John Wiley and Sons, Inc.

Servigny, A. d. and Renault, A. O., 2004. Measuring and Managing Credit Risk. New York: McGraw Hill

Companies, Inc.

Shimko, D.C. and Went, P., Credit Risk Management, The GARP Risk Series. Jersey City: Global Association of

Risk Professionals.

Stephanou, C. and Carlos M. J., 2005. Credit Risk Measurement under Basel II: An Overview and Implementation

Issues for Developing Countries. World Bank Policy Research Working Paper Series.

Topp, R. and Perl, R., 2010. Through-the-Cycle Ratings versus Point-in-Time Ratings and Implications of the Mapping between Both rating Types. Financial Markets, Institutions and Instruments, Vol. 19, Issue 1, pp.47-61.,

Treacy, W. F. and Carey, M.,2000. Credit risk rating systems at large US banks. Journal of Banking and Finance,

vol.24(1-2), pp.167-201.

US Department of the Treasury 2001. Comptroller Handbook, rating credit risk. New York: office of the comptroller

of currency.

Varotto, S., 2008. An Assessment of the Internal Rating Based Approach in Basel II. Journal of Risk Model

validation, Forthcoming.